Achieving a better match between ordered and actual performance of urban low-noise asphalts – exploring better solutions for noise abatement

Simon Steiner 1

Swiss Federal O ffi ce for the Environment FOEN, Noise and NIR, Road Noise, Bern, Switzerland

Felix Schlatter 2

Grolimund + Partner AG - environmental engineering, R&D division Thunstrasse 101a, 3006 Bern

Erik Bühlmann 3

Grolimund + Partner AG - environmental engineering, R&D division Thunstrasse 101a, 3006 Bern

ABSTRACT Thanks to the paradigm shift in noise abatement, over 1000 low-noise road surfaces have been constructed in urban areas in Switzerland in the last ten years. Until now, such road surfaces were often ordered according to a Swiss standard for semi-dense asphalts (SDA). The di ffi culty with low noise road surfaces in urban areas from a noise control point of view is that the acoustic performance often varies by more than 3-4 dB when newly constructed – a variability in product that is hardly acceptable to noise abatement professionals. This study investigates how a better match can be achieved between the ordered and the actual e ff ect, by narrowing down the mixture design in the Swiss standard for SDA. For this purpose, several variants for a revised standard were analysed regarding the statistical risk of acoustic non-conformity. The analysis is based on a large dataset from a connected InterNoise22 paper where statistical analyses of acoustic measurement data and data from the laboratory tests of over 200 low-noise surfaces in Switzerland are presented. This paper makes recommendations for revised mixtures in an improved Swiss standard for SDA, carefully balancing the advantages of the di ff erent variants with the practical limitations they bring.

1. INTRODUCTION

Noise pollution is a global problem and road noise is by far the largest source [1]. In Switzerland, for example, during daytime one in seven people and one in eight at night are a ff ected by harmful or annoying road tra ffi c at their place of residence [2]. To protect humans and nature, the Swiss

1 simon.steiner@bafu.admin.ch

2 felix.schlatter@grolimund-partner.ch

3 erik.buelmann@grolimund-partner.ch

a slaty. inter.noise 21-24 AUGUST SCOTTISH EVENT CAMPUS O ¥, ? GLASGOW

Environmental Protection Act (EPA) of 1983 [3] stipulates that environmental pollution control measures should preferably be carried out at source. Regarding road noise, protection at the source consists of the following options: Low-noise road surfaces, speed reduction, quiet tyres and restrictions on use. A next pillar of the EPA is the polluter pays principle, which applies also to the road noise. This means that the owner is responsible for complying with the applicable environmental limit values of his installation, in our case the road infrastructure. Accordingly, the owner must ensure that his installation complies with the provisions of the Noise Abatement Ordinance (NAO) [4]. It is therefore imperative that the noise abatement measures employed achieve their intended e ff ect. The Figure 1 shows this circumstance in relation to the noise protection measure low-noise road surfaces.

Quality control

Noise problem

Subsidises (FOEN)

Limit value exceedances; Measures required

Noise problem solved

Offers, evaluation and contracting

Requirements & specifications

Delivery and installation Quality control

Payment

Road surface must be replaced at an early stage

SDA 4 with 3dB Long-term effect

Mostly at the expense of the owner

Figure 1: Schematic flowchart of a noise remediation measure.

After a noise problem has been identified, the road must be rehabilitated at the owner’s expense, while receiving between 16% and 32% in federal subsidies [5], if the required noise reduction is achieved and proven by means of tyre / road noise measurements [6]. In this context, the Federal O ffi ce for the Environment (FOEN) supports the 26 cantons as well as cities and municipalities in Switzerland with the protection of the population from excessive noise by promoting low-noise road surfaces by subsidising the installation of low-noise road surfaces. This financial support will continue in the future as the protection of people from noise is regarded as a permanent task. When using low- noise road surfaces as a noise abatement measure, a noise reduction e ff ect of -3 dB (in the case of SDA 4) and of -1 dB (in the case of SDA 8) is generally required at the end of its service life to qualify for subsidies. This noise reduction refers to the Swiss noise model StL-86 + [7, 8] this reference value corresponds approximately to a conventional dense asphalt (mix of DAC 11 / SMA 11 at 5-10 years of age). Consequently, it is of great interest that the noise reduction measure used delivers what it promises, causing otherwise problems for the road owner. However, experience with the installation of semi-dense asphalts according to the Swiss standard SNR640:436:2015 [9] with maximum aggregate size of 4 mm has shown that the desired e ff ect cannot always be achieved. The following Figure 2 shows the initial noise reduction when the low-noise surface are installed. The figure shows that there is a wide range of acoustic quality: For 4mm surfaces, a span of 3-4 decibels is not uncommon. The same is also true for 8mm road surfaces. The consequences of a poorly performing low-noise surface, apart from the increased noise pollution for the residents, are primarily to be borne by the owner, whose installation does still not

30

Agg. size [mm]

25

4 8

20

n [-]

15

10

5

0

−8 −7 −6 −5 −4 −3 −2 L CPX, 8% N2 [dB]

Figure 2: Distribution of noise reduction after installation of the low-noise surfaces according to the Swiss standard for SDA.

comply with the regulations. Practically, this means that the road surface has to be replaced at an early stage [6]. Consequently, it is of great interest that the noise measures used, deliver what they promise. This paper discusses the current practice of installing low-noise road surfaces in urban areas, following the Swiss standard for semi-dense asphalts. The paper also shows that the probability of success of a road surface can be significantly increased with an adjustment of the Swiss standard for low noise surfaces. With the work of this paper, not only the cost of the road infrastructure for the owner can be reduced, but the population can also be better protected from excessive and annoying noise.

2. THE HISTORY OF LOW NOISE ASPHALTS IN URBAN AREAS IN SWITZERLAND

In recent decades, several programmes and research projects have been carried out in Switzerland to improve mixture design, adapt manufacture and construction methods with the overall goal to optimise the durability and acoustic and mechanical properties of low noise road surfaces.

2.1. Early research in urban low-noise surfaces (2003-2011) In Switzerland, the research on low noise road surfaces in urban areas at lower speeds has started in the early 2000s by investigating existing solutions from tests performed in Switzerland, and international literature [10]. It was found that road surfaces achieved good acoustical properties with maximum aggregate sizes of 4 mm with void volumes between 6-14% [11, 12].

2.2. The focus on SDA as a compromise between acoustics and durability (2012) Inspired by these promising results, the FOEN and the Federal O ffi ce for Roads (FEDRO) decided to finance a research package on low-noise pavements in urban areas [13]. Within this research package, the best products of low-noise road surface types in research were evaluated and 15 test tracks were constructed. Until the year 2017, the surface texture, as well as the acoustic behaviour of the tracks was monitored in detail [14]. These test road surfaces were constructed according the first version of the Swiss standard for SDA published in 2013 (SNR 640 436:2013) and consisted of di ff erent semi-dense and porous road surfaces with void contents between 6 and 22 %. The monitoring results have indicated that the road surfaces with maximum aggregate size of 4 and 6 mm have good acoustic performances of up to 8 dB quieter than the reference surface [14].

2.3. Minor adjustments (2015) Based on the first findings of the research package in 2015, the most recent version of the Swiss standard [9] was published. The main change in the standard is in the naming of classes of the SDA mixtures and the lowest void content class with 8 % target void content was removed (which actually belongs to the class of conventional dense rolled asphalts). This yields to SDA road surfaces defined in the void range between 10-22% (SDA 4) and 10-18% (SDA 8). The standard includes the following void content classes for SDA 4 and SDA 8:

– -12: Target void content in mixture 12 %, accepted range between 10 to 14% – -16: Target void content in mixture 16 %, accepted range between 14 to 18% – -20: Target void content in mixture 20 %, accepted range between 18 to 22% (only SDA 4)

2.4. Wide scale application as low noise road surfaces (2012-2022) After the publication of the standard in 2013, it was possible to benefit from the experience of low- noise road surfaces in urban areas. Since that time, the installation of low-noise road surfaces as a noise protection measure has greatly increased, as suggested by Figure 3. Note that this figure only contains the road surfaces, on which acoustic measurements (using the CPX method [15]) by Grolimund + Partner AG have been performed. As not all the road surfaces are monitored by this company, the total number of installed road surfaces tends to be slightly higher than reported here. Thus, since 2012 a total of over 1000 low-noise road surfaces have been gradually been installed and monitored over all parts of Switzerland. Detailed information has been collected from 200 of these road surfaces on the mixture design and drill core investigations [16]. The location of these road surfaces with detailed information on mixture composition and acoustics is shown in the lower right corner of Figure 3. This dataset has been used in a connected InterNoise22 paper [16], in which statistical analysis on the construction as well as acoustic parameters has been performed. The current paper discusses the implications of a new mixture design on low noise road surfaces.

2.5. Current installation practice The following Figure 4 shows the relationship between the mixture design and the void content for SDA 4 and SDA 8 for the total 200 road surfaces. The graph shows the di ff erent standard variants depending on the void content according to the current standard. Hereby, the standard allows for the construction of di ff erent void classes (i.e SDA 4-12: Void content of the mixture in the range of 10 to 14%). The grading curves in the figure clearly show that the standard is formulated very broadly and that large parts accepted by the standard are not used in practice at all. With a broadly formulated standard, a wide variety of mixtures are possible, resulting in di ff erent acoustic results. Therefore, part of the large scatter in the initial values when newly constructed is due to the broad formulation of the standard. Conversely, a broad standard enables the promotion of innovation. The problem, however, is that usually only compliance with the standard is required, without defining criteria for acoustic performance. The colour di ff erentiation of the grading curves shows how the void contents depends on the grading curve. Higher proportions of the sieve fraction 1, 2 and 4 mm (for 8 mm road surfaces) lead to lower void content. The e ff ect of higher void content is illustrated in the last row of Table 1, in which the measured initial average acoustic noise reduction to the reference model (StL-86 + ) is shown.

2.6. Long term assessment Experience with the low-noise road surfaces shows that the acoustic quality of the road surfaces decreases over time [17]. The following Figure 5 shows the relationship between di ff erent current

Figure 3: Distribution of low noise road surfaces in Switzerland shown for each year of installation since 2012. The lower right map displays all the 200 low noise road surfaces, which are included in this paper and thus have detailed information on construction parameters as well as CPX measurements. Please note that not all the road surfaces installed, are depicted in this figure (see text for explanation).

Table 1: Average grading curve for di ff erent variants of SDA 4 & SDA 8. The table shows the average value per void content class ( ± 2%) and the average noise reduction ( ∆ L CPX ) in comparison to the Swiss reference model StL-86 + . (Approximately a mix of DAC 11 and SMA 11 at 5 to 10 years of age)

SDA 4-12 SDA 4-16 SDA 4-20 SDA 8-12 SDA 8-16

Sieve 0.063mm 9.1 (1.5) 9.3 (1.9) 7.9 (1.7) 7.2 (1.0) 6.4 (0.7)

Sieve 0.5mm 15.7 (1.3) 13.2 (1.5) 11.4 (1.2) 11.8 (1.1) 10.0 (0.7)

Sieve 1mm 20.0 (2.2) 15.7 (2.3) 13.4 (1.7) 14.9 (1.6) 12.7 (0.7)

Sieve 2mm 29.5 (4.4) 23.2 (4.8) 22.1 (5.1) 20.7 (2.1) 17.6 (2.0)

Sieve 4mm 92.4 (4.0) 91.4 (4.6) 94.4 (2.7) 90.6 (6.7) 29.4 (5.4)

∆ L CPX -5.9 (0.9) -6.4 (0.7) -7.4 (0.6) -4.4 (1.0) -5.2 (0.9)

norm variants in void classes ( ± 2 %) and the acoustic ageing. It clearly illustrates that the initial acoustic value is highly dependent on the void content. Accordingly, the best acoustic values are

—— Low noise surfaces * Detailed infos available

100

SDA types

SDA types

4 mm agg. size

8 mm agg. size

90

SDA4-12 SDA4-16 SDA4-20

SDA8-12 SDA8-16

80

70

Mass passing [m-%]

Max sieve curve SNR 640 436:2015

60

Max sieve curve SNR 640 436:2015

50

Max sieve curve VSS 40 436:2021

Max sieve curve VSS 40 436:2021

40

(proposal)

(proposal)

30

20

10

Min sieve curve SNR 640 436:2015

0

11.2

0.063

0.125

0.25

0.5

1.0

2.0

4.0

5.6

8.0

0.063

0.125

0.25

0.5

1.0

2.0

4.0

5.6

8.0

Sieve [mm]

Sieve [mm]

Figure 4: Average grading curves for the subtypes of SDA 4 (left) and SDA 8 (right) along with their standard deviation. The di ff erent void contents are graded in colour. ( ± 2%)

measured at high void contents, whereas low void content lead to higher initial noise levels.

−1

78

SDA4-12 SDA4-16 SDA4-20

SDA4-12 SDA4-16 SDA4-20

70

L CPX, mixed traffic (8% heavies) [dB]

−2

76

68

−3

L CPX 2000 Hz [dB]

L CPX 315 Hz [dB]

−4

74

66

−5

64

72

−6

SDA4-12 SDA4-16 SDA4-20

62

−7

70

−8

60

0 1 2 3 4 5 6 7 Age [a]

0 1 2 3 4 5 6 7 Age [a]

0 1 2 3 4 5 6 7 8 Age [a]

Figure 5: Acoustic long-term performance of SDA 4 mixtures in function of void content classes specified in the Swiss standard for SDA based on the statistics of 136 road surfaces (SDA 4 - 12 → max. aggregate size 4 mm, void content 12 ± 2%).

The figure also shows that the ageing behaviour of the road surfaces constructed after the Swiss standard varies in function of the target void content class. The e ff ects of void content and other mixture design and construction parameters on acoustic performance are examined in more detail in the related InterNoise22 paper [18]. While the ageing rate tends to be lower with low void contents, the road surfaces with higher void contents show a slightly higher ageing rate, but all road surfaces are subject to acoustic ageing e ff ects.

With increasing road surface age, the average acoustic noise reduction tends to converge and reach a certain saturation, currently between -2 and -3 dB. There is only an overlap of the ageing curves from about 6 years onwards. However, it must be taken into account that the data basis in this region is still rather thin and therefore not entirely representative. SDA road surfaces are assumed to have a service life of 10 to 15 years. Thus, it now appears that the road surfaces with a higher void content show the better noise reduction e ff ects in the first stage of life (0-6 years). How the road surfaces will behave in the future, however, requires further monitoring of the road surfaces. Earlier work suggests, for instance at least for 8 mm road surfaces that low void surfaces do not necessarily end better. [17]

79

−1

SDA8-12 SDA8-16

SDA8-12 SDA8-16

SDA8-12 SDA8-16

L CPX, mixed traffic (8% heavies) [dB]

68

78

−2

67

77

L CPX 2000 Hz [dB]

L CPX 315 Hz [dB]

−3

76

66

−4

75

65

74

−5

64

73

−6

63

0 1 2 3 4 5 Age [a]

0 1 2 3 4 5 Age [a]

0 1 2 3 4 5 6 7 8 Age [a]

Figure 6: Acoustic long-term performance of SDA 8 mixtures according the standard VSS 640:436- 2015 in function of void content based on the statistics of 64 road surfaces (SDA 8 - 12 → max. aggregate size 8 mm, void content 12 ± 2%).

If, on the other hand, the spectral signature of the ageing behaviour is considered, di ff erent noise generation mechanisms are to be distinguished. For instance, the low-frequency range of the CPX measurement can be associated with vibration sound, which in turn is an indicator of deteriorating texture. [19] On the other hand, higher frequencies indicate a functioning pore network with communicating and from the surface accessible pores. Regarding the low frequency range, there is practically a linear relationship between age and the measured noise level. Only towards the end of the data series the trend seems to break slowly and flatten out asymptomatically. This again shows that the layers with the most cavities tend to have the highest levels towards the end of the data basis. At high frequencies, it can be seen that road surfaces with low void content have significantly worse starting values than surfaces with a higher void content. These road surfaces also have voids, but it seems that they are not equally e ff ective because the connection between the voids is probably not given. Consequently, in order to benefit from the positive e ff ects of voids, the road surface must have a certain level of voids ( > 12%). A deeper and more thorough analysis with regard to void content is presented in the connected paper [16]. It concludes that road surfaces with starting values lower than -6 dB at the beginning of their service life seem to have a connected pore network, helping to reduce noise in the long-term. With increasing void content, however, there is an elevated risk that the acoustic benefit can be lost, which in turn is reflected in the somewhat higher ageing rates. This is because the interconnected pores, which are responsible for the good performance, gradually become clogged and thus inactive. Thus, it can be observed that the surfaces with the highest void content end at a comparable noise

level as the most dense surfaces start at the beginning of their life.

2.7. Recommendations for revised mixtures Based on the considerations presented in the previous chapter regarding long-term noise reductions of di ff erent void classes and the thorough analysis performed in the connected paper [16, 20] , a proposal for a revised mixture design is formulated below in the following Table 2. The table shows a proposal for the statistically optimal mixtures with a void class of 13 - 17 % and for comparison, the current limiting norm curve (SN 640 436:2015) is also listed. The table also contains a proposal for a newer version of the standard (SN 40 436: 2021, dark grey columns), which is currently in the consultation phase.

Table 2: Proposal of optimal mixture design for SDA4 road surfaces.

Proposal SNR640 SN 640 SNR640 VSS SN 40

Mixture 436:2015 436:2021 436:2015 436:2021

Parameter

(Draft) (Draft)

Name SDA4-15 SDA4-12 SDA4-12 SDA4-16 SDA4-16

Marshall void [%] 13..17 10..14 10..14 14..18 14..18

Sieve 0.063 mm 6.5..(8.2)..10* 3..12 3..11 3..12 3..11

Sieve 0.5 mm 10..16 4..24 4..21 4..24 4..21

Sieve 1 mm 12..20 7..29 7..25 7..29 7..25

Sieve 2 mm 18..(27)..30* 12..50 12..35 12..50 12..35

Sieve 4 mm 90..100 90..100 90..100 90..100 90..100

Acoustic factor [21] < 42** - 47*** - - - -

Drill core voids (single value) 12..18% 10..20% 14..24%

Drill core void content (av.) 13..17% 10..18% 14..22%

Degree of compaction 99 ..102% ≥ 98% ≥ 98%

Binder content >= 6.0 % >= 6.0 %

*Upper limit, when rest of sieve curve is reduced, i.e sieve fraction 2.0 mm

**Optimal conditions

***Extended range.

In the table, some areas have been listed as extended areas. This is because it has been shown that good results can be achieved by carefully selecting the sieve fractions in di ff erent combinations [16]. The fact that the interaction of di ff erent sieve fractions is decisive has already been discussed in earlier work [21]. In this work, it was found that the interaction between the fine fractions (Filler) and the sand fractions is decisive and the so called Acoustic Factor has been developed. The connected paper, however, includes the recommendation to define a new void class (SDA 4-15) with a void content between 13 and 17%. Furthermore, the proposal also includes the recommendation to implement a stricter limit value for the degree of compaction of an SDA. The compaction has a statistical influence on the long-term performance of the road surfaces, as for undercompacted road surfaces the risk of earlier failure is elevated.

3. CHALLENGES AND BENEFITS OF NARROWER MIXTURE DESIGNS FOR SDA

3.1. Perspective noise abatement At the beginning of this paper, it was shown that the initial noise reduction e ff ect of the installed low-noise road surface is subject to a large variation. For e ff ective noise protection, however, it is of interest to have solutions for eliminating the noise problem that work as long as possible. With the proposed adjustment of the grading curves, there is now a much narrower range of road surfaces that can be installed. This should generally reduce the scatter in the initial noise reductions, when installing low-noise road surfaces. Figure 7 and 8 illustrate the statistical success rates of di ff erent variants for SDA 4 respectively SDA 8 road surfaces at the beginning of their service life in function of their void content (i.e -12: 10- 14%). The success rate is defined for 4 mm road surfaces according the Swiss definition for low-noise road surfaces [22], which states that promising road surfaces should reach an initial noise reduction of at least -6 dB (SDA 4) in comparison to the Swiss reference noise model StL-86 + .

Promising noise reduction

Promising noise reduction

Promising noise reduction

56.6 %

65.5 %

72.4 %

SDA 4-12

SDA 4-16

SDA 4-optim.

Total: 53

Total: 55

Total: 29

27.6 %

34.5 %

43.4 %

Unsatisfactory noise reduction

Unsatisfactory noise reduction

Unsatisfactory noise reduction

Figure 7: Success probabilities for SDA 4 road surfaces, expressed as the percentage of achieving at least -6 dB initial reduction for di ff erent void contents (Void content ± 2%).

Promising noise reduction

Promising noise reduction

Promising noise reduction

60.0 %

71.4 %

80.0 %

SDA 8-12

SDA 8-16

SDA 8-optim.

Total: 40

Total: 15

Total: 14

20.0 %

28.6 %

40.0 %

Unsatisfactory noise reduction

Unsatisfactory noise reduction

Unsatisfactory noise reduction

Figure 8: Success probabilities for SDA 8 road surfaces, expressed as the percentage of achieving at least -4 dB initial reduction for di ff erent void contents (Void content ± 2%).

In an analysis of the entire data set, it was examined how the noise reduction of the proposed mixture develops over time compared to all other known SDA road surfaces. As the proposed mixture is intended to ensure noise reduction that also lasts over a longer period of time, there must be a balance between good initial e ff ect and long-term e ff ect. Therefore, the compromise of an increased void content was deliberately chosen in order to e ff ectively benefit from the connected pores at the beginning, whilst not being exposed to an elevated ageing rate due to increased void content. For this

purpose, the following Figure 9 shows the ageing behaviour of the proposed mixture design compared to all other road surfaces.

6 4 mm road surface

21

2 8 mm road surface

9

−1

−2

19

4

39

19 15

−3

−2

65

L CPX, 8% Heavies [a]

L CPX, 8% Heavies [a]

69

33

5

−4

52

−3

12

86

10 6

55

118

−5

15 18

119

−4

14

20

−6

Optimised All Pavs.

Optimised All Pavs.

−5

14

29 29

0 1 2 3 4 5 6 7 8 Age [a]

0 1 2 3 4 5 6 7 8 Age [a]

(a) SDA 4

(b) SDA 8

Figure 9: Temporal development of the mixture-optimised road surfaces in comparison to all other road surfaces. The numbers in the graph represent the number of measurements.

This shows that the adapted mixture for the SDA 4 road surfaces (at least in the initial phase) can provide significantly better noise reductions (Figure 9a). The exponential extrapolation of the trend (dashed line) also shows that these road surfaces develop slightly better. However, this trend is subject to great uncertainty as there is still relatively little data. This is even more pronounced for the SDA 8 road surfaces (Figure 9b). The figure also shows that the SDA 4 surfaces average between -2 and -3 at the end of the data series. For SDA 8, this value is in the range of -1 to -2 dB. This, however, is only an estimation based on the first half of the expected service life of the road surfaces, as the expected service life is between 10 - 15 years.

4. CONCLUSIONS AND RECOMMENDATIONS

This paper illustrated that low-noise road surface can be used e ff ectively for noise abatement and the 4 mm road surfaces have a good potential for noise protection. Thereby, the initial installation noise reduction e ff ect is 1-2 dB higher than for 8 mm road surfaces. Therefore, the installation of SDA 4 in urban areas is advantageous for noise protection. However, it has been shown that not every mixture can achieve the same noise reduction e ff ect leading to a large variation in initial noise reduction of the road surface. A possible solution is discussed in this paper. Thus for a functioning noise protection, an adjustment of the grading curves in the underlying standard would be advantageous. This would allow to benefit from the most functioning areas of this type of road surface. That, the standard allows for di ff erent mixture designs and optimisation potential for low noise road surfaces is already discussed in previous work [16, 21]. With the optimisation of the grading curve the noise reduction e ff ect can be ensured. Thereby the success rate for optimal surfaces is higher for the SDA 4 than for SDA 8 as the data basis for SDA 4 was significantly larger and consequently more robust than for the SDA 8.

The long-term e ff ect of low-noise road surfaces has been discussed in previous work and is a crucial parameter for the success of low noise road surfaces [17]. The evaluations in this paper have shown that the noise reduction of road surfaces of di ff erent void content classes tend to be in a comparable range at the end of the data series. However, data covering the entire life cycle of low-noise road surfaces (up to 15 years) is still lacking and accordingly no scientifically substantiated statements can be made yet. In the first half of life, however, it becomes apparent that surfaces with an increased void content achieve the better noise reductions on average. However, slightly higher levels are indicated in the low frequency range of the road surfaces with higher void content, indicating texture-related damage. How these road surfaces evolve with time is subject to further investigation. The proposed optimised mixture is in the middle, between the void-rich and the more dense surfaces. The aim is to benefit from the durability of the dense surfaces without losing too much of the good initial properties of high void content. The evaluations of the long-term behaviour already indicated that a good compromise was found with the optimised mixtures. From the the perspective of noise abatement, the risk of unsatisfactory acoustic performance is significantly reduced with additional potential for better long-term performance. However, to prove the assumptions and the recommendation it is absolutely necessary to further investigate and monitor low-noise road surfaces also in the future. Thus, it is important to gather information on the evolution of the low noise road surfaces in the second half of the lifetime.

REFERENCES

[1] World Health Organization WHO. ENVIRONMENTAL NOISE GUIDELINES for the European Region . 2018. ISBN 978 92 890 5356 3. [2] Umwelt Bafu. Lärmbelastung in der Schweiz Lärmbelastung in der Schweiz. 2018. [3] Swiss Federal Council. Federal Act on the Protection of the Environment (EPA), 1983. URL

https://www.fedlex.admin.ch/eli/cc/1984/1122_1122_1122/en . [4] Swiss Federal Council. Noise abatement Ordinance (NAO). URL https://www.fedlex.adm in.ch/eli/cc/1987/338_338_338/en . [5] Bundesamt für Umwelt BAFU. Handbuch Programmvereinbarungen im Umweltbereich 2020 – 2024. Mitteilung des BAFU als Vollzugsbehörde an Gesuchsteller. Umwelt-Vollzug , 1817:294, 2018. URL www.bafu.admin.ch/uv-1817-d . [6] Gregor Schguanin and Toni Ziegler. Leitfaden Strassenlärm. Vollzugshilfe für die Sanierung. Stand: Dezember 2006. Umwelt-Vollzug Nr. 0637, (In German) Implementation aid for road noise improvement. Technical report, Bundesamt für Umwelt, Bern, 2006. URL https: //www.bafu.admin.ch/bafu/de/home/themen/laerm/publikationen-studien/pub likationen/leitfaden-strassenlaerm.html . [7] BUwAL. COMPUTERMODELL ZUR BERECHNUNG VON STRASSENLÄRM, Teil 1 Bedienungsanleitung zum Computerprogramm StL-86. Schriftenreihe Umwelt SRU , (60), 1987. URL https://www.bafu.admin.ch/bafu/de/home/themen/laerm/publikationen-s tudien/publikationen/computermodell-berechnung-strassenlaerm.html . [8] Bundesamt für Umwelt, Wald und Landschaft (BUWAL). Strassenlärm : Korrekturen zum Strassenlärmmodell. MITTEILUNGEN ZUR LÄRMSCHUTZ-VERORDNUNG (LSV) , (6):7, 2010. [9] SNR 640 436:2015. Semidichtes Mischgut und Deckschichten, Festlegungen, Anforderungen und Konzeption, 2013. [10] Hansueli Pestalozzi, Hans-Jörg Grolimund, and Christian Angst. Lärmarme Strassenbeläge innerorts Statusbericht 2003. Technical report, 2004. [11] Christian Angst, Françoise Beltzung, Dieter Bosshardt, Hans-Jörg Grolimund, and Hansueli

Pestalozzi. Lärmarme Strassenbeläge innerorts Schlussbericht 2007. Technical report, Bundesamt für Umwelt und Bundesamt für Strassen„ Bern, 2007. [12] Christian Angst, Francoise Beltzung, Dieter Bosshardt, Toni Ziegler, and Erik Bühlmann. Lärmarme Strassenbeläge innerorts Jahresbericht 2010. pages 1–56, 2010. URL www.astr a.admin.ch . [13] Jacques Perret, Jean-Pierre Bolli, and Thomas Arn. Research Package: Low-noise Pavements in Urban Areas: Report synthesis. (November 2017), 2017. [14] Erik Bühlmann, Philip Bürgisser, Toni Ziegler, Christian Angst, and Thomas Beckenbauer. Forschungspaket lärmarme Beläge Innerorts - Teilprojekt (TP) 3: Langzeitmonitoring. Technical report, 2017. [15] ISO 11819-2. Acoustics - Measurement of the influence of road surfaces on tra ffi c noise - Part 2: The close-proximity method (ISO 11819-2:2017). pages 2–4, 2017. [16] Erik Bühlmann and Felix Schlatter. Recommendations for the successful design of high performance urban low-noise asphalts – findings from the statistics of 200 mixtures. In Proceedings of INTER-NOISE 2022 International Congress and Exposition of Noise Control Engineering, Glasgow, Schottland , 2022. [17] Emanuel Hammer, Simon Steiner, Marlène Dias, and Erik Bühlmann. Long-term acoustical performance of low noise road surfaces in urban areas in Switzerland. In Proceedings of Euronoise 2015, Maastricht , pages 1315–1320, 2015. [18] Erik Bühlmann, Marlène Dias, and Simon Steiner. Influence of environment- and tra ffi c- related factors on acoustic ageing of low-noise road surfaces in Switzerland. In Proceedings of Euronoise 2015, Maastricht , pages 1321–1326, 2015. [19] Ulf Sandberg and Jerzy Ejsmont. Tyre / road noise. reference book. 2002. [20] Felix Schlatter, Daniel Schweizer, and Erik Bühlmann. Präzisierung Ausführungsbestimmungen Akustik SDA-Beläge. 2022. [21] Erik Bühlmann and Emanuel Hammer. Towardssemi-dense asphalt mixtures that guarantee acoustic performance and durability. INTER-NOISE 2017 - 46th International Congress and Exposition on Noise Control Engineering: Hong-Kong , 2017-Janua:1559–1568, 2017. [22] VSS. VSS-640425 Lärmmindernde Decken - Grundlagen, 2013.